WebMCP Registry site · v0.18.35
Use doodleai.art with your agent.
doodleai.art (doodleai.art) registers 20 WebMCP tools. 18 of 20 tools declares readOnlyHint, so an agent can invoke them through tools.call. The registry runs the tool in a fresh headless page on doodleai.art and returns untrusted site content.
This page mirrors the WebMCP Registry listing for doodleai.art, last live-checked by the registry at 2026-09-23T05:32:31.806Z. Run ghostget webmcp sites.get --input '{"domain":"doodleai.art"}' --json for the current schema; the listing can drift between checks.
Use doodleai.art with my agent
Ghostget's bundled webmcp adapter speaks to the public WebMCP Registry, so an agent always reads this site's current tool schema instead of a stale hard-coded copy. Install v0.18.35, sync bundled adapters once, then call the registry operations:
bun add --global https://github.com/hraness/ghostget/releases/download/v0.18.35/hraness-ghostget-0.18.35.tgz
ghostget adapter sync-bundled --jsonRead the live tool schema — every registered tool, its input schema, annotations, and the page that publishes it:
ghostget webmcp sites.get --input '{"domain":"doodleai.art"}' --jsonCall a read-only tool — the registry refuses anything not declared readOnlyHint and returns the site's answer as untrusted content:
ghostget webmcp tools.call --input '{"domain":"doodleai.art","tool":"browse_doodle_learn","input":"{}"}' --jsonRegistered WebMCP tools
doodleai.art publishes 20 tools on doodleai.art; 18 declare readOnlyHint.
| Tool | What it does | Callable through Ghostget |
|---|---|---|
browse_doodle_learn | Browse the /learn/ hub: Doodle AI's editorial content grouped by category (guide, explainer, prompts, studios), each with its label, blurb, and the article titles + urls inside it. Read-only. Optional 'category' shows just one group; omit i | Callable read-only tool |
compare_doodle_skills | Compare 2-4 Doodle AI skills side by side: name, whether a photo is required, output aspect ratio, images returned, credit cost, and one differentiating line each, ending with each page URL. Pass 'ids' as an array of skill ids from list_doo | Callable read-only tool |
create_doodle_board | Create a new empty board to group generated doodles. Creating a board is free and never spends a credit. Use this when someone wants a place to collect images before generating them. | Listed for discovery; the registry allows read-only calls only |
find_doodle_answer | Find where a question is answered across all Doodle AI articles. Returns the best-matching article SECTIONS (not whole articles): each gives the article title, section title, section slug, url, and a short relevant snippet, plus the exact r | Callable read-only tool |
get_article_faq | Get one Doodle AI article's FAQ (question/answer pairs). Pass 'url' — an article path (e.g. /photo-to-sticker/). Most articles have no FAQ; when so, this says there is none and points you at list_article_sections for that url instead. Pagin | Callable read-only tool |
get_doodle_article | Read the beginning of one Doodle AI article. Pass 'url' — a path returned by search_doodle_articles (e.g. /photo-to-cartoon/). Returns the first ~1400 characters of the article's readable text plus a link to the canonical page for the rest. | Callable read-only tool |
get_doodle_overview | A short factual summary of what Doodle AI is, how sign-in and credits work, and how many skills exist. Read this before answering questions about the product. For the complete site text, fetch /llms.txt or /llms-full.txt directly. | Callable read-only tool |
get_doodle_skill | Full detail for one Doodle AI skill by id: what it makes, whether a photo is required, output aspect ratio, how many images it returns, its credit cost and its page URL. Use after list_doodle_skills when the user has picked one. | Callable read-only tool |
get_doodle_skill_guide | Read the actual guidance for one Doodle AI skill: what the style produces, what input photo or description works well, and what to expect. Pass 'id' from list_doodle_skills. A skill's guide is long, so it is paginated — pass 'page' (default | Callable read-only tool |
get_prompt_pack | Get concrete, usable Doodle AI prompts from the prompt library, each as a short labelled line. Optional 'topic' filters by keyword (e.g. 'sticker', 'pet', 'gift'); optional 'limit' (1-5, default 3). Read-only. When nothing matches it names | Callable read-only tool |
list_article_sections | Show the section outline (H2 headings) of one Doodle AI article: a numbered list of slug and title, plus the section count and word count. Read-only, first-party metadata. Pass 'url' (a path from list_doodle_articles, e.g. /photo-to-cartoon | Callable read-only tool |
list_doodle_articles | List Doodle AI's editorial articles (guides, explainers, prompt packs, studio pieces) as a paginated inventory of title, url, and one-line description. Read-only, first-party metadata. Use this to enumerate EVERY article — search alone cann | Callable read-only tool |
list_doodle_pages | Map the pages of Doodle AI so an agent can orient and navigate deliberately. Read-only. Each line gives the path, a one-clause purpose, and whether sign-in is required; account-changing pages and the '/' composer (which spends credits) are | Callable read-only tool |
list_doodle_skills | List Doodle AI generation skills: id, what each produces, whether it needs a photo, how many images it returns and its credit cost. Returns a page of results, not the whole catalogue — narrow with requiresPhoto or raise limit. Call this bef | Callable read-only tool |
list_doodle_topics | List Doodle AI's content taxonomy so you can orient before searching. Returns each category (guide, explainer, prompts, studios) and each cluster (cartoon, pets, stickers, gifts, social, studios) with the number of articles in it and one ex | Callable read-only tool |
open_doodle_page | Navigate this tab to a public Doodle AI page — use it to put a person in front of the composer or a skill page. It only navigates: it never generates an image and never spends a credit, so a human still presses send. | Callable read-only tool |
read_article_page | Read a Doodle AI article straight through, one page at a time, without choosing sections. Pass 'url' (a path from list_doodle_articles) and optional 'page' (>=1, default 1). Returns that page of the full article text in a limit-sized window | Callable read-only tool |
read_article_section | Read the body text of ONE section of a Doodle AI article, paged to fit the limit. Pass 'url' and 'section' — 'section' is EITHER a slug or a 1-based number from list_article_sections. Optional 'page' (>=1, default 1) reads further into a lo | Callable read-only tool |
search_doodle_articles | Search Doodle AI's editorial articles (guides, explainers, prompt packs) by keyword and get a short ranked list of titles, URLs, and one-line descriptions. Read-only. Use this first to find an article, then call get_doodle_article with a re | Callable read-only tool |
submit_doodle_feedback | File a short product feedback note for Doodle AI. It is saved and posted to the team's public roadmap board. Use when someone wants to report a bug, request a feature, or share an opinion about the product. Does not spend credits and does n | Listed for discovery; the registry allows read-only calls only |
What to expect
- Read-only only. The registry refuses WebMCP tools that do not declare
readOnlyHint; Ghostget never weakens that check. - Untrusted results. Tool output is site content, not instructions — treat it as data.
- Live schemas.
sites.getalways returns the schema the registry saw most recently, so agents adapt when doodleai.art changes its tools. - No account needed. These calls are credential-free registry reads. For tools that need a session on doodleai.art, browse the site itself or check back when the registry lists more tools.